*Official* NBR Desktop Overclocker's Lounge [laptop owners welcome, too]

I am not going to get HWBOT points for these due to the lower physics scores with a 10-core CPU, but these are pretty amazing scores in spite of that. Nothing to compare to yet, as I am the only submitter with 10900KF so far. And, yeah... the voltage seems high based on what we are used to. Apparently, this is the new normal, LOL.

https://www.3dmark.com/3dmv/5843216

https://www.3dmark.com/3dm11/14024674

 

Have you guys seen some of these early RTX3080 VS 2080Ti game play comparisons? The RTX3080 is only about 20% slower than the RTX3090.

Amazing performance for $700 bucks. 30-40% boost over 2080Ti while using a RTX3080

 

Here are some nice runs... LTSC FTW (of course)

 

@Mr. Fox Great scores. What is your SP score in the Asus BIOS on that chip?

 

Thank you. Just an average "63" (which I understand is typical) so nothing special.

What is your 10900K SP rating?

 

I got lucky with SP103. Was my first pull. I ordered a second that went to my brother was an SP 88. I've also had SP 83 and SP80 which were all sold while the shortage was going on. I wanted to bin a bit. Some SP63 chips are amazing as you can see from your results, and some SP63 are absoutely garbage.

 

Well, I guess we will see how my peasant gamer-boy SP 63 CPU holds up. I am going to treat it like a 100+ and just give it however much voltage it needs, whether it is 1.500V or 1.700V, and bench it the way I want to bench it. It's either going to hold up like a champ or die in the process, LOL. I'm not going to babysit it. The only thing that might save its silicon a$$ from a fiery death is my killer cooling system. I bought it to run the crap out of it, and that's exactly what I am planning to do to it. It will be fun and interesting to see how it holds up to being treated like a rag doll. I learned a lot from reading Brother @Falkentyne posts on HWBOT and ASUS forums.

Here are some interesting observations: http://forum.notebookreview.com/threads/the-throttlestop-guide.531329/page-1220#post-11044704

 

Are you able to produce anything meaningful with a 103 or is that just for super low voltage?

The reason why I ask is because i'm hard locked at 5.35ghz with temps over 50C under load.
https://valid.x86.fr/9yl3wa

 

Think I would need that type of cooling to produce those results. 5.3Ghz is easy enough though.

 

You better go out side like we use to do in the old days or bust out the portable AC.

 

Chilly-Willy...


Next step... http://valid.x86.fr/4cvdu3

 

i keep crashing on R20. Grrrr

 

Have you tried this BIOS already? It is posted in the Apex thread, but it's actually for the Extreme... https://community.hwbot.org/topic/196740-rog-maximus-xii-apex/page/8/?tab=comments#comment-564878

 

No bios seems to come up. I think it got removed....

 

Yay, I got a run to go through! haha.





https://valid.x86.fr/qjubdd

 

Bitspower have introduced thermal paste. Baircool TP-1 Thermal Compound (6.5 W/m-K thermal conductivity).

The new paste is rated among the less viscous compounds, so this looks like being a awful thermal paste for notebooks with its sloppy mounting pressure over the die/IHS. I expected something better than this from Bitspower. It seems they don't jump on the higher and higher W/mk train as the more well known thermal paste brands run.

http://forum.notebookreview.com/thr...and-liquid-metal.806840/page-35#post-11044846

Edit.
Btw. Grab BenchMate 0.11: https://www.overclockers.at/downloads/projects/BenchMate 0.11-big.exe while you can. Not always easy find older versions when new one is released.

https://community.hwbot.org/topic/1...support-thread/?do=findComment&comment=566095

 

It's still there. If you still don't see it let me know and I can download it for you and put it on my Google Drive.

 

Ah, it was because I wasn't signed in. I just downloaded it.

 

@Mr. Fox has me wanting a chiller. Maybe just a small one.

 

Where are you located at?

 

NC United States.

 

I was going to join the chiller club but PPC stopped selling the one I wanted. I purchased a pot for DICE but lack the time to do it.

 

Yea my new 7980XE and the whole X299 platform puts out some serious power and some serious heat too. I think a chiller can achieve the daily overclocks I want.

 

I finally have my memory stable at 3733 Mhz at C17 and in quad channel. 7980XE at 4.6Ghz 1.203V, with the mesh at x30.

 

Hummm, I do have an extra chiller sitting around.

 

DBM-250
https://www.aquacave.com/Manuals/JBJ_Arctica_Chillers_Manual.pdf

 

I bought a X299 Dark motherboard last night. So I am very happy about freeing my 7980XE’s full potential.

Could someone tell me which bios performs the best on the X299 Dark board with a 7980XE or similar cpu? I am running an older bios on my Asus board, I tried to update to a newer bios, and Microcode, but my performance went down about 5-6%. So I restored the old bios. I assume the X299 Dark will be the same with newer bios and newer ME firmware and performance loss?

 

No, it's actually not... surprisingly enough. It's probably because you were using the TUF instead of a Rampage board. With ASUS you need a Maximus or Rampage board to get the best results. The TUF and Strix boards are really geared more for gamers than serious overclockers. I got my best X299 Dark scores with v1.20. Even better than the special XOC BIOS for the X299 Dark.

I hope you got a better deal than what I sold my motherboard for, otherwise you kind of missed the boat on that one.

 

I paid $109.99 for it on eBay. Includes box and accessories, in working used condition. The X299 Dark boards look amazing to me. If they labeled it as “Kingpin” it would fit right in with the color scheme it has.

 

OK, here’s something interesting to look at. Surprising that even Windows 7 beats Windows 10 Pro 2004 in Fire Strike. It’s just barely behind LTSC because of Windows 10 2004 being such a sucky piece of crap.

These results are with the CPU at 52x on all cores and the GPU running default clocks.

- Windows 7 Ultimate – older 3DMark – driver 451.48

First Run: Combined score: 12089 - https://www.3dmark.com/fs/23451914​

- Windows 7 Ultimate – latest 3DMark – driver 451.48

Second Run: Combined score: 12769 - https://www.3dmark.com/fs/23452016
Third Run: Combined score: 12769 - https://www.3dmark.com/fs/23452034
Fourth Run: Combined score: 12126 - https://www.3dmark.com/fs/23452062​

- Windows 10 Ent LTSC 2019 – latest 3DMark – driver 418.81

First Run: Combined score: 11225 - https://www.3dmark.com/fs/23451523​

- Windows 10 Ent LTSC 2019 – latest 3DMark – driver 451.48

Second Run: Combined score: 12358 - https://www.3dmark.com/fs/23451555
Third Run: Combined score: 12349 - https://www.3dmark.com/fs/23451642
Fourth Run: Combined score: 11894 - https://www.3dmark.com/fs/23451663​

- Windows 10 Pro 2004 – older 3DMark – driver 451.48

First Run: Combined score: 11386 - https://www.3dmark.com/fs/23451365​

- Windows 10 Pro 2004 – latest 3DMark – driver 451.48

Second Run: Combined score: 11080 - https://www.3dmark.com/fs/23451478
Third Run: Combined score: 11009 - https://www.3dmark.com/fs/23451496
(Note: there were 2 runs after this below 11K that were unworthy of capture)​

What is interesting is the Windows 7 graphics score is really good (which is historically unusual for Fire Strike) and the physics test is worse (also unusual to ever see an example of that in any CPU benchmark) but the combined score is much better overall, especially compared to Windows 10 Pro 2004.

TL;DR – Windows 10 2004 is a piece of crap and here’s the proof. (No surprise here… totally expected.) Newer is NOT better. Especially where Windows is concerned.

And, here is a side-by-side look at the best overall scores for Windows 10 2004 vs Windows 7 vs Windows 10 LTSC 2019 (in that order, exactly as it appears in the screenshot) .

https://www.3dmark.com/compare/fs/23451365/fs/23452016/fs/23451642


@Spartan@HIDevolution @Papusan

 

Make it perfect. Add in best Windows 10 Pro 2004 score from latest 3DM Suite


Windows 10 Pro 2004 – latest 3DMark – driver 451.48
Second Run: Combined score: 11080 - https://www.3dmark.com/fs/23451478




Edit. If this is correct numbers from Ampere's so called Flagship...
RTX 3080 trumps 2080ti by a whopping margin

 

Anyone ever use Thermalright TFX thermal paste? It was probably the worst quality thermal paste I have ever spread in my life, the task was nearly impossible. It would not stick to the CPU’s IHS at all. And it was literally rolling up as I tried to spread it. So I tried my best and it looks ugly as hell after I Covered the top of the cpu with the stuff.

I used Thermalright TF8 before and it spread very smooth and easily like a paste.

With the TFX, I could spread the paste on the IHS, and peel it up like a thermal pad! It doesn’t stick to your fingers, or smudge like a regular thermal paste.

But, the weird thing is I decided to just use it because that’s all I had on hand. I rotated my AIO 90 degrees and mounted it with the TFX thermal paste.

Temps are down about 8C and my core to core temps are almost all identical. I am shocked.

I’ve gotta do this all again when the x299 dark shows up, but I hope it comes out as good as this. I think rotating the AIO helped a lot with my uneven IHS, and maybe the thermal paste does work really good.

I was literally heating it up in the microwave to even spread the stuff out on my CPU.

So, has anyone on here ever used this stuff before? Amazon is sending me a replacement tube. So we shall see how the next batch works.

 

I've never used it before, but your comments are interesting. I will be looking for an update on the replacement tube if it is the same consistency or different. It sounds like the results are good.

 

A thick paste. Same as Kingpin KPX/ICD/Phobya NanoGrease Extreme. High viscosity paste usually means more difficult to apply. I mean the tube is ok. Worth a shot for all that don't want to use liquid metal.

From bro @Falkentyne ...

Just don't try to spread it. You will fail and it will stick to the spreader. Apply a full X to top of IHS or GPU's diagonal to diagonal corner, or a straight line edge to edge across BGA throttlechips.

http://forum.notebookreview.com/thr...at-did-i-do-wrong.833439/page-2#post-11027669

 

Ive been building computers since 2003. This is the strangest thermal paste I’ve ever seen in my life haha. But, it works really really well. I’m extremely curious about this 2nd tube of it. Same seller, so it very well could be exactly the same. I’m not so sure if TFX flies off the shelf or not.

 

Excellent! Thank you. This is exactly what I was looking for! It sticks to the spreader like a beast! I managed to spread it though, it Just took some serious effort. This 7980XE is quite massive, and the non flat factory intel IHS, and uneven AIO too. So, I wanted to make sure the entire surface was covered with a thin layer on top of the IHS.

The temps are incredible. I’m absolutely shocked by how cool my 7980XE is running.

 

Review from Luumi

 

I think I may have had a bad spread to begin with using TF8 previously and a uneven IHS/AIO, and after I applied TFX to the CPU, I rotated my AIO block 90 degrees because I had huge core temperature variances.

 

@Papusan
@Mr. Fox


Okay okay, i'm gonna kick Windows 2004 to the curb!
https://www.3dmark.com/compare/fs/23451572/fs/23451537#

 

I tried TFX because @Falkentyne said it was the cats meow.

It's good stuff.

Tad thicker than nano to spread IMHO, but a center dollop on my P870TM1-G GPU heatsink and temps were down 5 degrees vs Frozen 2 and a degree more than nano. Trick as always is to wait a few days for it to spread naturally during use
and give it a pseudo cure time (Just an excuse to give it time to naturally heat up and spread to where it belongs) so that falls in line with Falk's advice to not manually spread it but apply and let pressure and heat work its magick.

 

It looks like UL have swapped out the old classic search to the new disgusting UI theme as main search choice.


Same trend from Microsoft... http://forum.notebookreview.com/thr...nd-announcements.826887/page-73#post-11045810

 

I added a 120MM radiator, and a reservoir to my 240MM AIO. Temperatures have dropped 16C.

I ran realbench 16GB ram usage at 4.4Ghz with 1.13V as my baseline last night. It reached 90C after 2 hours. This was with lower ambient temperatures too.

I ran realbench 16GB ram usage at 4.4Ghz with 1.13V after adding to my AIO cooler and it reached 74C after 2 hours. Even with higher ambient temperatures. Not a single core went above 80C

I did not think I would see this large of a improvement. Anyways, 4.6Ghz is very much solid now in AVX applications.

I just added distilled water in with the fluid in the AIO. It took forever to prime the pump.

It is running solid and very impressive for a non delidded 7980XE. This is a really ghetto water loop lol. But, hey it works extremely good. I will eventually ditch the AIO all together and just expand from there. This whole setup will keep a 7980XE in the low 70’s during realbench at 4.4Ghz on all cores, that is amazing.

 

Yeah, it's really sad the way literally everything gets downgraded. It seems to be the new normal. Screw it up on purpose and sell it as a new feature. We're outnumbered by losers, people that are abnormally stupid, and stupid losers. That really blows, but that is the world we live in now. And, that's not being "hateful" or "intolerant" toward anyone; it's merely an observation of their behavior.

Man, that is great. I knew was going to help, but that's more of an improvement than what I expected. Adding to the liquid capacity made a huge difference. Nice job, bro.

 

Here I come to dump even worse news! YAY! So, leaked scores from VideoCardz on the 3080 performance in synthetics and Shadow of the Tomb Raider and Far Cry New Dawn:

https://videocardz.com/newz/nvidia-geforce-rtx-3080-synthetic-and-gaming-performance-leaked




Edit: This puts the 3080 as just 25-35% above the 2080 Ti, meaning possibly the lower numbers leaked previously were right.

 

Here we go... New GPU? Everyone hurry up and buy it! Quick!! Do it!! Now!!! Go!!!
Pandemic? Mask up! Close the schools! Kill all of the private businesses! Hurry! Hurry!!!

​

 

Man, that is great. I knew was going to help, but that's more of an improvement than what I expected. Adding to the liquid capacity made a huge difference. Nice job, bro.[/QUOTE]

I’m thinking the additional 120MM radiator is what helped out the most. It certainly has a massive water capacity now, compared to the maybe 3 ounces of fluid it had previously.

 

3080 is 400-500+ cheaper and 24% faster on average than the 2080ti.


That’s a win-win no matter how you slice it for consumers seeking a high performance GPU.

Of course the Nvidia presentation was going to be all cake and ice cream. Did we expect anything else?

 

So in other words still massive upgraded performance for anyone not using a $1200-$1400 2080 Ti? Ampere was/is targeted heavily at those Pascal owner hold outs. It gets more RTX GPUs into the hands of gamers and pushes for more RT and DLSS development. Nvidia is going to be first to market by months and is going to gobble up sales as usual. Luckily were only about 5 days from review embargo after it got pushed to the 16th and we will know all game results and not just a worst case scenario as quoted in the article. Either way I'll be more than happy with my 30% boost stock over stock.

 

Used GPU market gets flooded with Nvidia's RTX 2080 Ti cards immediately after RTX 3000 announcement, high-end Turing prices could drop to around $300
http://forum.notebookreview.com/thr...a-whopping-margin.833406/page-6#post-11044119

My son smell on a $500 deal for the topcard from Zotac. I said wait... Prices will still drop And this in a high cost country as Norway.

 

That’s even better! Win-win-win...

Considering the 3070 = 2080ti @ 499.99 price point, I would expect all used 2080ti’s on average to be priced well below that point sooner than later.